This clinical trial will test the efficacy of cytomegalovirus (CMV) glycoprotein B (gB) vaccine (Aventis Pasteur) with MF59 adjuvant (Chiron Vaccines) to prevent maternal CMV infection, using a phase II, randomized, placebo controlled, double-blind clinical trial. Healthy young women are screened for antibody to CMV on the post-partum wards of 4 hospitals. Those who are seronegative are invited to participate in the clinical trial. Participants are enrolled and immunized (3 doses of vaccine, 0, 1 and 6 month schedule). Participants have follow-up visits every three months for three years after the third dose of vaccine in order to screen them for CMV infection, collect samples for immunogenicity studies, identify pregnancies and births, and collect safety information. All infants born to clinical trial participants are tested for congenital CMV infection by virus culture at birth. As of August 31, 2003, over 12,000 women had been screened for antibody to CMV, and 252 subjects enrolled. Twenty-eight CMV infections have occurred so far, a rate of 6.6% per year. There were 78 pregnancies among 66 trial participants, and 2 congenital CMV infections among 54 live births (3.7%). Important accomplishments in this trial to date include: enrolled around two-thirds of planned sample, developed a novel, accurate and inexpensive means of endpoint (CMV infection) detection for CMV vaccine clinical trials, recruited a study population with a CMV infection rate that allows us to accumulate the largest group of healthy asymptomatic young women with primary CMV infection ever studied, established use of immunologic and virologic assays that will allow us to learn a great deal from this trial whether or not vaccine is effective in preventing maternal infections. Specific aims will test the ability of CMV gB/MF59 vaccine to prevent maternal infection, evaluate the rate of congenital CMV infection in offspring of trial participants, determine whether maternal CMV infection is associated with miscarriage, characterize the virologic and immunologic features of primary CMV infections, evaluate novel approaches to end-point identification in CMV vaccine clinical trials, define long-term immunogenicity of study vaccine (antibody to CMV gB, neutralizing antibody and cell mediated immune response) and compare rates of adverse events as well as local and systemic reactions in CMV gB/MF59 recipients to placebo recipients. Based on enrollment to date and the rate of CMV infection in trial participants, we expect to have 60 to 90 CMV infections in the study population and approximately 200 pregnancies. With planned enrollment and these event rates, the clinical trial will achieve the power needed to test efficacy for prevention of maternal infection and will provide a preliminary estimate of whether immunization affects rates of congenital CMV infection.